The automotive industry continues to undergo massive changes and transformations that are having a profound impact on our world – on the ways we live, work, travel (and conceptualize travel), and more. In my eyes, the innovation impacting the automotive industry is one of the most interesting, most important topics in modern business. Many of these changes are coming in the areas of simulation, machine learning, data analytics, artificial intelligence (AI), and high-performance computing (HPC) – Altair’s core areas of expertise. 

Recently, I sat down for a Q&A with DataNet, an IT magazine based in South Korea. In the interview, I shared my thoughts on many of the most vital changes shaping the automotive landscape, and what technologies are driving these changes. In summation, I believe it is an exciting time to be working in AI in the automotive industry. Since the earliest days of the automotive industry, there have been very few periods as momentous as the one we are in now. Truly, the industry is transforming in front of our eyes. Increasing the use of data, AI, and HPC is an opportunity that courageous, efficient companies will capitalize on to stand atop the industry for years to come. 

***

Q: Please introduce yourself and your role at Altair.

Andrew Baldwin: My name is Andrew Baldwin, and I am the global vice president of sales for AI and automotive at Altair. In my role, I am the sales leader for AI focused on supporting Altair’s customers that are global automotive OEMs and suppliers. I began my career in data software in 2012 as an intern at a company called Datawatch, which was acquired by Altair in 2018. Throughout my career, I have worked in many different industries including banking, government, and retail manufacturing; I am now focused exclusively on automotive. I have previously led teams in both North America and the Europe-Middle East (EMEA) region when I lived in London, England. During my career, I have focused on bringing value to customers through the strategic application of data analytics, machine learning, and AI.

Q: What is the overall level of AI adoption in the global automotive market?

AB: One of the best parts of working in the automotive space is the fast-paced adoption of AI. It seems that every week there are exciting new developments surrounding AI in the automotive industry. That said, the global market is still only scratching the surface of what is possible with AI and machine learning. There are so many areas that AI can be and has been applied that it would take too long to describe them all here, but I will highlight a few of the key areas of AI in the automotive industry below. 

Predictive Maintenance: AI in the automotive industry is being used to predict when vehicles need maintenance. This not only prevents surprise repairs and downtime for the driver, it also extends the lifespan of components in the vehicle. The same principles can also be applied to the heavy assets and machinery used to create and assemble vehicles and components. While AI is being used widely for predictive maintenance today, I think we will continue to see wide expansion both in the number of models using AI and the number of different components AI is being used to predict. In the future, I expect to see broad use of AI in the automotive industry in the areas of battery health and state of charge, especially in e-vehicles, along with other more basic maintenance like air filters, headlights, etc.   

Manufacturing Efficiency: With tight margins and a competitive landscape, the automotive industry will continue to invest in AI to improve manufacturing efficiency. The value of improving first time through rate, reducing rework, and scrap is clear but the application of AI in these areas is not yet mature. Using machine learning techniques such as decision trees allows companies to optimize their designs and production process based on material availability with a high degree of accuracy.  

At Altair we focus on AI democratization. In other words, we enable key people like process engineers to leverage AI without needing to become expert coders. The value of making AI available to employees who are not data scientists comes in being able to get more powerful models into production quickly. Models that used to take six months to develop can now be developed in a week. The saved time allows engineers to spend more time improving their products.   

Supply Chain: The complexity of the automotive supply chain has long created challenges for OEMs and suppliers – these challenges now present opportunities for the application of AI in the automotive industry. One good example is the difficulty in accurately determining safety stock quantities of all required components and, in turn, providing accurate demand to the global supply chain. This is further complicated by fluctuations in market demand, changing customer preferences, materials shortages, and the performance of logistics carriers. 

For many auto manufacturers, no centralized data analysis process is used to consider potential risk and demand variability and then create a recommendation for optimal order quantities. Instead, many companies rely on experience-based assumptions from their production planners to determine optimal inventory and safety stock levels. Not only is this method error-prone, but it also prevents an analysis of each individual part number, as typically thousands of part numbers are involved in the vehicle manufacturing process. 

Altair has pioneered an innovative solution leveraging AI/machine learning to address these inventory optimization challenges. AI/machine learning algorithms excel in handling complex and big data, recognizing patterns, and adapting to changing conditions. Altair’s solution optimizes safety stock calculations and minimizes stock-related risks while improving overall inventory quality. This drives costs down, improves customer satisfaction, and gives Altair’s customers a competitive edge.

The Altair® RapidMiner® platform inventory optimization solution:

• Uses predictive analytics to predict order quantities
• Uses prescriptive analytics to recommend the optimal order quantities
• Optimizes the weekly order schedule to minimize weekly costs
• Fully automates the process, eliminating the need for operator oversight

I anticipate many automotive companies adopting similar approaches to inventory optimization and other supply chain opportunities soon. 

Energy: Using AI to optimize fuel efficiency in internal combustion engine (ICE) vehicles is another area that has been done but will continue to improve. With market pressure on these combustion engine vehicles using fossil fuels as effectively as possible is critical, especially as drivers continue to become more cost and environmentally conscious as it relates to fuel. 

Even more than ICE engines, electric vehicles (EVs) stand to gain enhanced performance from AI. Many of the drawbacks to EVs are energy and battery related. Being able to accurately predict range given a specific set of circumstances is important to the driver. But more broadly, AI and data can be used to efficiently allocate and build charging stations to ensure proper coverage for EV owners.    

Overall, I see AI being used broadly in the automotive industry, but its depth and effectiveness will continue to grow quickly in the coming months and years. OEMs and suppliers that invest early and intelligently in AI will gain a large advantage over their competition.    

Q: What value does data have in the automotive industry? Please introduce some success stories from the automotive industry that use data well.

AB: The value of data and AI within the automotive industry is nearly limitless. The vast amount of historical data – coupled with the amount of new data gathered every day – gives organizations a huge set of information to build and train models on. 

A great example of this is in warranty analytics. We collaborated with JLR to create a warranty analysis tool. This tool allows JLR to bridge the organizational silos that have traditionally limited the value of data within automotive companies. With Altair, they can share data and the insights they get from that data across their organization.  

This tool is a machine learning model that automatically categorizes warranty issues and assigns them to the engineering team for implementation, using the customer and technician data received from dealers during a warranty repair. Prior to using the Altair RapidMiner platform and Altair’s data science expertise, JLR had minimal success auto categorizing the issues and were forced to use their quality engineers and domain experts to manually review the issues. 

The results from Altair’s solution are impressive – the model automatically categorized issues with an accuracy level of more than 90%. This is a significant improvement over the baseline, and JLR reported increased warranty cost savings, more accurate warranty-related financial predictions, faster issue identification, and improved satisfaction. On top of all that, JLR’s engineers and domain experts were able to focus on what they do best and stop generating endless pivot tables. 

This is just one example of AI in the automotive industry, but it illustrates the transformative power of data and how it can deliver meaningful results to your business.  

Q: Are there any technologies or trends that are gaining attention in the global market?

AB: Of course. Here are some of the major technologies and trends that I see:

Electrification: The continuing shift towards EVs is one of the most important trends in the automotive industry. EVs are more environmentally friendly than gasoline-powered vehicles, and they are becoming more affordable.

Connectivity: Connected cars are vehicles that are equipped with internet connectivity. This allows them to communicate with other vehicles, infrastructure, and even the driver's smartphone. Connected cars offer a variety of benefits, such as improved safety, better navigation, and personalized services.

Sustainability: The automotive industry is under increasing pressure to become more sustainable. This is leading to the development of new technologies that can reduce the environmental impact of vehicles, such as lightweight materials and energy-efficient engines.

Manufacturing advancements: The automotive industry is also investing in new manufacturing technologies, such as 3D printing and robotics. These technologies can improve efficiency and productivity, and they can also be used to create new and innovative designs.

Mobility solutions: The traditional car ownership model is being challenged by new mobility solutions, such as ride-hailing and car sharing. These services offer a more flexible and more affordable way to get around, and they are becoming increasingly popular.

Autonomous driving: Autonomous driving is another major trend in the automotive industry. Self-driving cars have the potential to revolutionize transportation, making it safer, more efficient, and more convenient.

Q: What value does Altair aim to provide in the manufacturing industry, and what is Altair's strategy for achieving this?

AB: Altair is a global technology company with partners in every industry and talented people in offices all over the world. After nearly 40 years in the business, we know the importance of investing in research and development to ensure we bring the best innovations to our customers. Whether that’s through our patented intellectual property or our strategic acquisitions, we continue to lead with our cutting-edge technology, and plan to do so for the next 40 years and beyond.

At Altair, we take great pride in leveraging our technology across industries. This allows us to deliver the best products to our customers at the best possible price. In fact, many of our data analytics techniques were built in and battle tested within the financial services space. Accelerated innovation happens when these ideas and techniques shift from one industry to another. 

For example, both our best-in-class decision trees and our unique ability to connect to and visualize streaming data in true real-time were originally developed for financial services customers. However, in the last four years Altair has taken these cutting-edge data strategies and applied them to manufacturing. This makes sense as Altair got our start in automotive. In fact, every global automotive OEM and Tier 1 supplier uses Altair’s innovative products and solutions. And seven of the top 10 automotive Fortune 500 companies use our Altair RapidMiner platform for data analytics.

We strive to be the trusted partner in reducing “friction” within organizations and implementing AI and data to gain a competitive edge. Altair can do this regardless of where an organization is on their data journey since Altair RapidMiner, our data analytics and AI platform, can help overcome the most challenging obstacles in the way. We offer a path to modernization for established data analytics teams as well as a path to automation for teams just getting started. We do this without requiring your organization to radically change your people, processes, computing environment, or existing data landscape, helping you achieve your data goals without fundamentally changing who you are or what you have. 

Altair RapidMiner:

• Scales AI initiatives without requiring a big team of data scientists or expensive services engagements. Organizations can upskill their workforce so users from novice to expert can leverage the tools needed to provide data-driven insights.  
• Empowers users to extract and prep data easily from any source, working with reports and PDFs that are core to the business. 
• Alleviates the pressure of modernizing expensive legacy environments. Teams can create, maintain, and run SAS language programs, models, and workflows directly in a multi-language environment (e.g. Python, R, SQL).
• Gets more models into production. Teams can operationalize models faster and monitor them continuously across one shared MLOps environment.
• Processes and displays massive amounts of fast-changing data. Users can build sub-second streaming, batch, and business intelligence (BI) data applications.

There are several key differentiators for Altair, but I’d like to focus on two. The first enables us to accelerate deployment of our solutions, while the second delivers world-class solutions at a fraction of the cost. 

First, is the accelerated deployment of AI. This is a critical part of our strategy as many of our customers have had experience with open-source or other vendors where it takes a long time to get just a small percentage of developed models into production. We can accelerate this through our open solution philosophy. We integrate into our customers’ existing ecosystem, working with their existing tools, regardless of what those tools are. Our goal is to help our customers become more efficient, and we can’t do that if we require them to rip out the parts of their system that are working – so we don’t. We work with them. And that allows our customers to get up and running faster.  

The second is our proprietary customer-focused units licensing model. Altair Units are a concurrent, all-inclusive licensing model. This means, instead of buying seats for 150 different products, Altair customers buy a level of concurrent usage or “bandwidth” and get access to the entire suite. Each user “checks out” a specific product (or products) that withdraws a specific number of units from the pool; when the user logs out of the software, the units are returned to the pool so others can use them. This allows for the same Altair Units pool to be used, for example, by users in the U.S., then subsequently by users in Japan. Or it allows the same user to use several Altair products in parallel without needing separate licenses. You could think of this as similar to Netflix: lots of content for a fixed cost (just without any login restrictions!).

The Altair Units model stands in contrast to the consumption-based model. Consumption-based models are easy to start but become very expensive at scale. The named-user model is also popular, but it’s limited because each license is tied to a specific user. Altair’s proprietary licensing model allows shared and on-demand access to more than 150 Altair and partner software products, without adding incremental cost. This is especially beneficial for situations where you rarely access a tool, and can’t justify the cost of a license, but really need access from time to time. Altair gives you the ability to do just that. Our customers report that Altair Units save, on average, 30-50% license cost compared to traditional models.